What would it take, in terms of the incoming data between now and the December meeting, to convince the FOMC to raise rates for the third time this year?
The key data, of course, are for core inflation, following the sharp and now persistent slowing in core PCE inflation and core CPI inflation after February this year. The principal driver of that decision will be the FOMC’s assessment of the underlying inflation rate at that point. The answer may well depend on which measure of inflation they use to judge the underlying inflation rate.
The Committee usually judges the underlying inflation rate to be the 12-month core PCE inflation rate. Looking at the monthly data on a year-over-year basis is especially appropriate for interpreting inflation data, given that there’s some evidence of residual seasonality. Plus, over shorter time horizons, the signal relative to the noise is considerably smaller.
But this is an unusual situation: The 12-month rate that will be available at the December meeting might not be sufficiently informed about the underlying rate of inflation. The slowdown in core inflation began with a sharp decline in the price level of wireless communications in March. Such a price level shock has no implications for the medium-term inflation outlook, but its effect on the 12-month rate will persist until that shock passes through that 12-month window. Arithmetically, the price level shock alone lowered the 12-
month core PCE inflation rate by a couple of tenths, at least on the surface accounting for most of the slowdown in the 12-month rate.
Dudley emphasized this point in a recent interview:
The reason why inflation won’t get up to 2% very quickly on a year-over-year basis is that we’ve had these very low inflation readings over the last 4 or 5 months. So it’s going to take time for those to sort of drop out of the year-over-year calculation. But I think if you look at inflation sequentially, in other words, what’s inflation likely to actually do over the next 6 months, I’m expecting somewhat higher readings than what we’ve had over the last 6 months.
Still, while all participants appreciate the price level shock, many, if not all, are nevertheless uncertain about whether this was the sole source of the slowing in core inflation. Yellen said, for example, that “there could be more going on there,” and Dudley said, “the jury is still out.” So, a “wait and see” posture is definitely called for.
So, let’s consider some scenarios for the inflation readings that the FOMC will have in hand at the December meeting. The first and the easiest is that the annualized rate of core PCE inflation going forward, that is from July to October, is 1.5%, and there are no more clearly identifiable price level shocks to account for this. With that assumption, the 12-month rate through October would be 1.4%. In that case, the FOMC would judge the underlying inflation rate to be around 1½%, too low to warrant a hike in December. That outcome for core inflation would be terrible! Given the flatness of the Phillips curve, the FOMC might not, in this case, be able to achieve its 2% inflation objective over the medium term.
Our second case is that the annualized rate of core PCE inflation going forward is 1¾%, which the FOMC for some time believed to be the underlying rate. That would be a relief. Let’s look at what the 12-month rate would be at that point: still only 1.5%. But the annualized rate over the period following the March shock, from April through October, would be 1.7%. While that’s an uncomfortably low 12-month rate, we believe that would be a go for a rate hike because of the performance of inflation over the shorter window after the March shock. And it would be a fairly easy decision, though it would likely be contested. After all, in June the median dot for 2017 implied the third hike this year, and the median projection for Q4/Q4 core PCE inflation in 2017 was 1.7%, consistent with a core PCE inflation rate of 1.8% after the March shock. And the Committee will take into account, in assessing the 12-month rate, that past price shocks are still depressing it.
A third case is that core inflation turns out to be 1.9% in future readings. That wouldn’t be too surprising because that was the 12-month rate for most of the period from August 2016 to February 2017. If 1.9% were judged to be the underlying rate, the FOMC at that point would have virtually achieved its inflation objective. (As we like to say, “good enough for government work!”) In this scenario, the inflation rate after the March shock through October would be 1.8%, good enough to hike even though the 12-month rate would still be only 1.5%. Again, remember that the March shock will still be depressing the 12-month rate at this point (as it will into 2018).
There is, unfortunately, one more case. What if core PCE inflation going forward is only 1.5%, but there are further identifiable price level shocks? Still a no-go. Caution is called for.
The 12-month and Shorter-horizon Core PCE Inflation Rates in Alternative Scenarios
|Future core PCE inflation (July to October)||Core PCE inflation from April to October||12-month core PCE inflation rate through October||December rate hike?|
|1.50% (with or without any additional identifiable price level shocks)||1.53%||1.37%||No|
We remain comfortable with our call that there will be a hike in December, though admittedly a bit less so since the softer readings for core inflation. We see the shorter-horizon inflation rate picking up enough to move the FOMC to raise rates in December.